5 ms·

Since 2001, the potential size of a Kernza seed has doubled, and scientists hope to lengthen its productive lifespan from five to 10 years. That's pretty fast

by jobu 9y ago

Since 2001, the potential size of a Kernza seed has doubled, and scientists hope to lengthen its productive lifespan from five to 10 years. That's pretty fast for traditional selective breeding techniques. It would be interesting what kind of progress could be made with genetic engineering. Seems like they're trying to avoid the GMO label, so I would guess it's another 15-20 years before Kernza will be ready for widespread agricultural use.

DubiousPusher 9y ago
Just speculating here but that rapid improvement may be because it was a less popular cultivar and thus wasn't exposed to as much tinkering. What I mean is perhaps no one came along and picked the low hanging fruit until now.
vanderZwan 9y ago
> That's pretty fast for traditional selective breeding techniques. I wouldn't be surprised those techniques have seen a lot of improvements in recent years too. I mean, surely genetic engineering coming onto the scene doesn't mean everyone just dropped the other methods? There must be a lot of complementary techniques there - I'm sure a lot of modern advances in biology are helpful there as well. EDIT: Heh, I just got to the part of the article where it explicitly describes this: > Kernza’s DNA has never been genetically engineered; its genes get reshuffled via the scattering of pollen, from the male parts of flowers onto the feathery, sticky female parts. But advances in genomics—the -sequencing of DNA—over the last 15 years have made it far easier to tweak Kernza. Almost all of the grain’s genome has now been mapped. Once breeders have a genetic blueprint, they can track down the genes that control particular traits and select individuals with genetic stock that codes for, say, fat seeds or resistance to disease. I hope they're smart enough to keep some kind of "unselected seed bank" to keep a source of higher genetic diversity for quick adaptation.
- Obi_Juan_Kenobi 9y ago
  There's marker assisted breeding, QTL analysis, etc., but the hard limit is generation time. If you want to introgress a particular trait into a useful background, it's going to take ~5 generations to do it adequately. That generally means 5 years, or sometimes you can squeeze in a winter nursery in the southern hemisphere. Proper QTL analysis requires making suitable inbreds (several generations of selfing), creation of the RILs (another several generations of selfing after the making the F2s), and finally a couple years for gathering trait data. Oh, and this all requires generation of proper markers and so forth. Starting from scratch, this is a decade-long process if everything goes well. As for diversity, you collect wild accessions and those are generally maintained by a seed co-op, often something associated with a university.
  - vanderZwan 9y ago
    Thanks for the enlightening follow-up!

by jobu 9y ago

DubiousPusher 9y ago

Just speculating here but that rapid improvement may be because it was a less popular cultivar and thus wasn't exposed to as much tinkering. What I mean is perhaps no one came along and picked the low hanging fruit until now.

vanderZwan 9y ago

> That's pretty fast for traditional selective breeding techniques. I wouldn't be surprised those techniques have seen a lot of improvements in recent years too. I mean, surely genetic engineering coming onto the scene doesn't mean everyone just dropped the other methods? There must be a lot of complementary techniques there - I'm sure a lot of modern advances in biology are helpful there as well. EDIT: Heh, I just got to the part of the article where it explicitly describes this: > Kernza’s DNA has never been genetically engineered; its genes get reshuffled via the scattering of pollen, from the male parts of flowers onto the feathery, sticky female parts. But advances in genomics—the -sequencing of DNA—over the last 15 years have made it far easier to tweak Kernza. Almost all of the grain’s genome has now been mapped. Once breeders have a genetic blueprint, they can track down the genes that control particular traits and select individuals with genetic stock that codes for, say, fat seeds or resistance to disease. I hope they're smart enough to keep some kind of "unselected seed bank" to keep a source of higher genetic diversity for quick adaptation.

Obi_Juan_Kenobi 9y ago
There's marker assisted breeding, QTL analysis, etc., but the hard limit is generation time. If you want to introgress a particular trait into a useful background, it's going to take ~5 generations to do it adequately. That generally means 5 years, or sometimes you can squeeze in a winter nursery in the southern hemisphere. Proper QTL analysis requires making suitable inbreds (several generations of selfing), creation of the RILs (another several generations of selfing after the making the F2s), and finally a couple years for gathering trait data. Oh, and this all requires generation of proper markers and so forth. Starting from scratch, this is a decade-long process if everything goes well. As for diversity, you collect wild accessions and those are generally maintained by a seed co-op, often something associated with a university.
- vanderZwan 9y ago
  Thanks for the enlightening follow-up!